Someone had to do it, cummins tubular manifold.

PASSENGER

Bells and Whistles
Joined
Jun 8, 2007
Messages
267
Just finished fabbing this up for a buddy, all equal length within 1/8". T4 split turbine, EMP coupling for each scroll as well as EGT for each scroll. Hope to fab up downpipe, turbo inlet and new intercooler pipe this weekend. All for a '07 24v.

cummins5.jpg

cummins4.jpg

cummins3.jpg

cummins2.jpg

cummins1.jpg
 
I take it thats for a common rail seeing as the mounting position for the turbo. But whats the big advantage of this over a porting and polishing a three piece?
 
Why would you want to have equal length with a turbo charged engine? Now, Naturally aspirated is a different story.
Enlighten me please.
Looks nice, the welds could be better looking though.
 
Yeah its for a common rail. The thing with LOG manifolds such as a 3 piece manifold is that they create excessive pressure, what I am trying to do get rid of the whole pressure (PSI) mentality and get going on a flow mentality. My experience with diesels comes from vw and mercedes diesels, I found that increasing flow, while decreasing pressure, gives more power, better fuel economy, lower egt's, lower peak cylinder pressures, etc.

Heres some of my vw work, these manifolds were on team rallyVW's TDI golf that just raced Pikes Peak a couple weekends ago, I believe they set a new record in their class.

lightmanifold.jpg

3rdmanifold.jpg
 
What do you use(material), and where do you get the short bent pieces? or do you bend them yourself?
THanks, TIm.
 
tony597fitter said:
Why would you want to have equal length with a turbo charged engine? Now, Naturally aspirated is a different story.
Enlighten me please.
Looks nice, the welds could be better looking though.


Ahhh but why do you say NA is so much different?

The reason equal length primaries are used is two fold:

Scavenging: This is what happens on a properly designed manifold in the collector. As the pulses arrive at the collector, and given the cylinder pairing was done correctly the pulses will arrive at EVEN INTERVALS, so as one pulse from one primary enters the collector it creates a vacuum in the direct neighboring primary and helps suck the pulse into the collector, this helps reduce the EMP (exhaust manifold pressure). Reducing exhaust manifold pressure helps reduce the amount of spent exhaust gasses that get trapped into the cylinder which prevents more fresh (and hopefully cold!) air re entering the cylinder on the intake stroke.

Pressure: Equal length primaries are also done to achieve a pressure balance on each individual primary, balancing the pressure on each primary also helps balance the egt for each primary, which helps balance the egt for each cylinder, and also helps balance the PEAK cylinder pressures across the cylinders. If there is more pressure on one primary than another the primary with more pressure keeps more spent exhaust inside that cylinder and creates the un desireable side affects mentioned above. The reason I used two EMP's (one for each side of the scroll) is in effort to monitor an imbalance between the cylinders. I can't of course expect them to be perfect yet as I still need to remove the head, remove the stock integrated cast "intake manifold" and build a real high quality lehman style intake manifold.

As for the comments on the welds its actually the coating that makes they look bad, its a ceramic coating that we applied too thick and it globbed funny on the ripples of my tig welds. I'll get some close up pictures of the welds for you guys to see, I am a tig welder by trade, high pressure pipe, SS sanitary welding etc.

The fitting on this manifold are sch40 1.5" (thats 4mm thick, or .157"). All the welds are full penetration, this manifold weighs in at 37 lbs. The flanges are CNC milled from 1018CRS .5" thick.

As for time in this manifold, well the construction was the fastest part, the engineering takes the longest, and all the programming for the flanges to run on the cnc mill, designing the collector and playing with angles, it all takes a lot of time, theres a lot of machining on this thing, the collector is cnc milled, the flanges are cnc milled, the straight sections are cut and beveled on a cnc lathe, and the fitting and welding was done by yours truly. Probably a good 200 hours into the project so far.
 
PASSENGER said:
Ahhh but why do you say NA is so much different?

The reason equal length primaries are used is two fold:

Scavenging: This is what happens on a properly designed manifold in the collector. As the pulses arrive at the collector, and given the cylinder pairing was done correctly the pulses will arrive at EVEN INTERVALS, so as one pulse from one primary enters the collector it creates a vacuum in the direct neighboring primary and helps suck the pulse into the collector, this helps reduce the EMP (exhaust manifold pressure). Reducing exhaust manifold pressure helps reduce the amount of spent exhaust gasses that get trapped into the cylinder which prevents more fresh (and hopefully cold!) air re entering the cylinder on the intake stroke.

Pressure: Equal length primaries are also done to achieve a pressure balance on each individual primary, balancing the pressure on each primary also helps balance the egt for each primary, which helps balance the egt for each cylinder, and also helps balance the PEAK cylinder pressures across the cylinders. If there is more pressure on one primary than another the primary with more pressure keeps more spent exhaust inside that cylinder and creates the un desireable side affects mentioned above. The reason I used two EMP's (one for each side of the scroll) is in effort to monitor an imbalance between the cylinders. I can't of course expect them to be perfect yet as I still need to remove the head, remove the stock integrated cast "intake manifold" and build a real high quality lehman style intake manifold.

As for the comments on the welds its actually the coating that makes they look bad, its a ceramic coating that we applied too thick and it globbed funny on the ripples of my tig welds. I'll get some close up pictures of the welds for you guys to see, I am a tig welder by trade, high pressure pipe, SS sanitary welding etc.

The fitting on this manifold are sch40 1.5" (thats 4mm thick, or .157"). All the welds are full penetration, this manifold weighs in at 37 lbs. The flanges are CNC milled from 1018CRS .5" thick.

As for time in this manifold, well the construction was the fastest part, the engineering takes the longest, and all the programming for the flanges to run on the cnc mill, designing the collector and playing with angles, it all takes a lot of time, theres a lot of machining on this thing, the collector is cnc milled, the flanges are cnc milled, the straight sections are cut and beveled on a cnc lathe, and the fitting and welding was done by yours truly. Probably a good 200 hours into the project so far.

Does all that equal a price comparable to say an ATS, BD, or Hightech manifold? Looks good and if you can log EGT, EMP, ETC numbers and show an improvement on current designs out there. I would sign up for one.:bow:
 
No the price would not be comparable, not even close, I just looked up on ATS' site and says their 03-06 with high temp coating is $600USD, this manifold would be over twice that much. This would be a you get what you pay for deal no doubt.
 
you make one up for a second gen and i will give you double that. defintely looks great and totally would help out egt wise. i am serious. i will buy one if your willling to make it.
 
Would this manifold not fit a second gen? I don't have any dodge knowledge, but the owner of this truck said it should fit a second gen no problem, I will be posting up fab pics of the progress this weekend so you guys can see how this looks installed in the engine bay. After we get it running and dyno'd and all that I'll probably be organizing a group buy on a few units. But testing first, not selling anything until I am %100 satisfied.
 
it will bolt right up to a second gen but the angle of the T4 flange is slightly different on the second gen. more angeled out of the fender well instead of straight down
 
Ok good to know, The reason I fit the turbo so close to the engine block is because the owner is running KORE suspension and it has really big shock towers that take up a lot of room. Oh and I should have my T4 to T3 adapter plates done tomorrow! I believe the turbo he is going to use is a s3100 or something like that, which is a t3 flange, but I couldn't make a collector small enough to fit just a T3 so I made it to T4 so he could run a t3 or a t4 turbo with the adapter.
 
How about strength? Most 24v headers I have seen have cracking problems.....
 
A few things went into the design for crack prevention, the outer stud holes are ovalized, the flanges are .5" 1018CRS, the primaries are 1020CRS .157 thick ( most exhaust headers you see on cars are 16 gauge, which is about .065") and its all full penetration tig welded.
 
Man that looks really good, big props on the design fabbin, I wouldn't even know where to start:hehe: , but seems like it would be real benificial to a real modded cummins
 
I have a question. When you say equal length are you referring to equal physical length or calculated equivalent length? Different numbers of different radius bends will result in different effective lengths, right? I am a neophyte as far as boosted engines go, so forgive me on this one. Isn't the pressure between the exhaust side of the cylinder head and the turbo supposed to be somewhat high to aid in spooling the turbo? You know what would trip my trigger is a less restrictive path on the intake side of the head. I like (and own) ZZ Fab's manifold, but I still have issues with how the air gets from the IC tube into the manifold. I guess what I'm getting at is I'd love to see something less of a plenum for an intake and something more on the order of a tubular manifold capable of delivering equal amounts of pressurised charge air to each cylinder. Maybe with finned tubes or some type of refrigeration evaporator tubing worked in to help further cool the charge air downstream of the IC?? I'm a design engineer with full 3D CAD capability, and I know people that could calculate the flows for me, but the world is an imperfect place. Truly testing for equal air delivery would have to be done using empirical data, like from a flow bench. Sorry, just lamenting....
 
Back
Top